PROJECT SUMMARY Persons with Alzheimer?s Disease and Related Dementias (ADRD) suffer from a 70% increase in mortality rates, frequent hospitalizations and medical complications, and catastrophic health care spending in the last 5 years of life. Recent policy efforts have focused on two important and related goals to improve ADRD care: supporting patients? preferences to remain in their own residence and avoiding potentially unnecessary and traumatic hospitalizations. However, performance on these objectives is suboptimal, with wide geographic variation. Virtually all of the prior research on the care and outcomes of patients with ADRD is derived from patients enrolled in traditional Medicare, with remarkably few studies of persons with ADRD in managed care plans. This raises a critical gap in knowledge because over the past 12 years, the number of beneficiaries enrolled in managed care (now called Medicare Advantage) more than tripled from 5.3 million in 2004 to 17.6 million in 2016. Under managed care, plans are paid a capitated rate to bear the risk of paying for covered services and are held accountable for quality of care. These incentives may yield innovative approaches to improve care and reduce costs for populations with ADRD. Alternatively, managed care plans may provide inadequate coverage of necessary services, avoid complex patients, or promote disenrollment among patients with high spending and intensive health care needs. The objective of this proposal is to evaluate the effects of the rise of Medicare Advantage on patient-centered outcomes in ADRD. Our working hypotheses are that, among persons with ADRD, enrollment in managed care will reduce the likelihood of long-term nursing home care, lower rates of hospitalizations and readmissions, and improve quality of care at the end of life, particularly among racial/ethnic minorities and plans that focus on dual-eligible patients. However, these effects will be attenuated after accounting for disenrollment from Medicare Advantage to traditional Medicare. Our specific aims are: Aim 1. Among patients with ADRD using home health services, examine the effects of Medicare Advantage on long-term nursing home care and acute hospitalizations; Aim 2. Evaluate the effects of Medicare Advantage on the outcomes of post-acute care among patients with ADRD; Aim 3. Evaluate the effects of Medicare Advantage on care at the end of life among patients with advanced dementia; and, Aim 4. Assess the heterogeneity of these effects across racial/ethnic minority populations and different types of Medicare Advantage plans. This project will use detailed comprehensive clinical and functional assessment data from all Medicare-certified home health providers and nursing homes in the US. Further, we will employ an innovative cross-temporal difference-in-differences approach that leverages the dramatic growth in Medicare Advantage enrollment from 2007 to 2017 to draw causal inferences. Our contribution is significant since federal policy continues to promote beneficiaries? enrollment in risk-bearing private plans without rigorous empirical evidence about the health consequences for vulnerable high-cost, high-need populations, such as those with ADRD.